Counterfactual confounding factor adjustment (COCOA) in single-cell RNA-seq matrix

0. Annotate cell types for all the cells

mkdir -p example/cfa

[ -f example/cfa/psych.annot.gz ] || \
    ../src/mmutil_annotate_col \
        --mtx data/brain_2018-05-03/filtered_count_matrix.mtx.gz \
        --row data/brain_2018-05-03/features.tsv.gz \
        --col data/brain_2018-05-03/barcodes.tsv.gz \
        --ann PsychENCODE.marker \
        --batch_size 1000 \
        --out example/cfa/psych

1. Extract cells with underlying phenotypes are annotated

meta.dt <- fread("data/brain_2018-05-03/filtered_column_metadata.txt.gz")
pheno.dt <- fread("data/brain_2018-05-03/phenotypes.csv")
model.file <- "example/cfa/pheno_model.rdata"

##########################################
## Propensity for pathological AD model ##
##########################################

if(file.exists(model.file)) {
    load(model.file)
} else {
    dir.create("example/cfa/", recursive = TRUE, showWarnings = FALSE)
    annot.dt <- .fread("example/cfa/psych.annot.gz")
    colnames(annot.dt) <- c("TAG", "celltype", "prob", "ln.prob")

    train.data <- pheno.dt[abs(pathoAD) <= 1 &
                           np_sqrt != -9 &
                           nft_sqrt != -9]

    prop.model <- glm(pathoAD ~ np_sqrt + nft_sqrt,
                      data = train.data,
                      family="binomial")

    save(list=c("prop.model"), file=model.file)
}

################################################
## these are cells with annotated individuals ##
################################################

.tag.file <- "example/cfa/input.tags.gz"
if(!file.exists(.tag.file)) {
    meta.pheno.dt <- merge(meta.dt, pheno.dt)[np_sqrt != -9 & nft_sqrt != -9][]
    .fwrite(meta.pheno.dt[, .(TAG)], .tag.file)
}

2. Create another file set for COCOA analysis

# Expected: input.{mtx,cols}.gz
# Column order may change
[ -f example/cfa/input.mtx.gz ] || \
    ../src/mmutil_select_col \
        data/brain_2018-05-03/filtered_count_matrix.mtx.gz \
        data/brain_2018-05-03/filtered_count_matrix.cols.gz \
        example/cfa/input.tags.gz \
        example/cfa/input

############################################################
## prepare for the input files matching the same order of ##
## cells/columns                                          ##
############################################################

.hdr <- "example/cfa/input"
.trt.file <- .hdr %&% ".trt.gz"
.ind.file <- .hdr %&% ".ind.gz"
.annot.file <- .hdr %&% ".annot.gz"
.lab.file <- .hdr %&% ".lab.gz"

if(any(!file.exists(.trt.file, .annot.file, .ind.file, .lab.file))) {
    cols.dt <- .fread("example/cfa/input.cols.gz") %>% 
        (function(x){ colnames(x)=c("TAG"); x }) %>% 
        left_join(meta.pheno.dt) %>%
        left_join(annot.dt)

    xx = cols.dt[, .(np_sqrt, nft_sqrt)]

    y.pheno <- data.table(y = predict(prop.model, xx)) %>%
        mutate(y = if_else(y > 0, "AD", "HC"))

    ind.dt <- cols.dt[, .(projid)] %>%
        mutate(y = unlist(y.pheno)) %>%
        mutate(ind = projid %&% "_" %&% y)

    .fwrite(ind.dt[, .(ind)], "example/cfa/input.ind.gz")
    .fwrite(y.pheno, "example/cfa/input.trt.gz")
    .fwrite(cols.dt[, .(TAG, celltype)], "example/cfa/input.annot.gz")
    .fwrite(cols.dt[, .(celltype)] %>% unique, "example/cfa/input.lab.gz")
}

3. Run the counterfactual adjustment model

# Just to have this in the same location
[ -f example/cfa/input.rows.gz ] || \
    cp data/brain_2018-05-03/filtered_gene_row_names.txt.gz \
       example/cfa/input.rows.gz

# Basic aggregate statistics
[ -f example/cfa/output.sum.gz ] || \
    ../src/mmutil_aggregate_col \
        --mtx example/cfa/input.mtx.gz \
        --col example/cfa/input.cols.gz \
        --annot example/cfa/input.annot.gz \
        --lab example/cfa/input.lab.gz \
        --ind example/cfa/input.ind.gz \
        --out example/cfa/output

# CF-adjusted aggregate statistics
[ -f example/cfa/output.resid_mu.gz ] || \
    ../src/mmutil_cfa_col \
        --mtx example/cfa/input.mtx.gz \
        --col example/cfa/input.cols.gz \
        --annot example/cfa/input.annot.gz \
        --trt example/cfa/input.trt.gz \
        --lab example/cfa/input.lab.gz \
        --ind example/cfa/input.ind.gz \
        --knn 10 --rank 30 --bilink 100 --log_scale --normalize \
        --out example/cfa/output

MIN.TOT.CUTOFF <- 50

.data.file <- "example/cfa/stat_tab.rdata"
if(!file.exists(.data.file)) {
    stat.tab <- .read.stat.all("example/cfa/output", tot.cutoff = MIN.TOT.CUTOFF)
    save(stat.tab, file = .data.file)
} else {
    load(.data.file)
}

stat.avg.tab <- .take.average(stat.tab, "AD", tot.cutoff = MIN.TOT.CUTOFF)

First, show confounders and genes associated with the confounding effects

.mat <- stat.tab[tot > 0, .(mu=sum(cf.mu)), by=.(gene, col)] %>%
    spread(key=col, value=mu, fill=0)

.ln.mat <- log(1 + .mat[, -"gene"]) %>% as.matrix %>% t %>% scale
.ln.mat[is.na(.ln.mat)] <- 0

gene.cf.assoc.dt <- test.gene.association(.ln.mat) %>%
    na.omit %>%
    rename(cf.beta = beta, cf.se = se, cf.pv = p.val) %>%
    as.data.table

top.genes <- 
    gene.cf.assoc.dt[order(gene.cf.assoc.dt$cf.pv),
                     head(.SD, 3),
                     by = .(pheno, celltype)] %>%
    filter(cf.pv < 5e-4) %>% 
    select(gene) %>% unique %>% unlist

top.cf.assoc.dt <- gene.cf.assoc.dt[gene %in% top.genes] %>% 
    na.omit %>% 
    mutate(row = gene) %>%
    mutate(col = pheno %&% "@" %&% celltype) %>%
    mutate(weight = -log10(pmax(cf.pv, 1e-8))) %>% 
    order.pair(ret.tab = TRUE) %>%
    as.data.table

## Loading required package: cba

## Loading required package: proxy

## 
## Attaching package: 'proxy'

## The following objects are masked from 'package:stats':
## 
##     as.dist, dist

## The following object is masked from 'package:base':
## 
##     as.matrix

plt <-
    .gg.plot(top.cf.assoc.dt, aes(y=row, x=col, fill=celltype, size=weight)) +
    theme(title = element_text(size=8)) +
    theme(axis.text.x = element_blank()) +
    theme(axis.ticks.x = element_blank()) +
    theme(axis.title = element_blank()) +
    ggtitle("Top " %&% length(top.genes) %&% " genes ~ confounding effects") +
    facet_grid(.~pheno, scales="free", space="free") +
    geom_point(pch=22, stroke=.1) +
    scale_fill_brewer(palette = "Paired") +
    scale_size_continuous("p-value",
                          range=c(0,3),
                          labels = function(x) num.sci(10^(-x)))

.file <- fig.dir %&% "/Fig_Confounding_Genes.pdf"
.gg.save(filename = .file, plot = plt, width=5, height=5)
print(plt)

PDF

.row.df <- match.row.with.pheno(.ln.mat)

.tsne <- Rtsne::Rtsne(.ln.mat, n_threads=16)
colnames(.tsne$Y) <- "tSNE" %&% 1:ncol(.tsne$Y)

.tsne.dt <- cbind(.row.df, .tsne$Y) %>% as.data.table

p1 <- .gg.plot(.tsne.dt, aes(x=tSNE1, y=tSNE2)) +
    theme(legend.position=c(1,1), legend.justification = c(1,1)) +
    geom_point(aes(fill=celltype), alpha=.8, stroke=.2, size=1.25, pch=21) +
    scale_fill_brewer("", palette = "Paired")

p2 <- .gg.plot(.tsne.dt %>% arrange(desc(disease)), aes(x=tSNE1, y=tSNE2)) +
    theme(legend.position=c(1,1), legend.justification = c(1,1)) +
    geom_point(aes(fill=disease), alpha=.5, stroke=.2, size=1.25, pch=21) +
    scale_fill_manual("", values=c("#FF9999","#FFFFFF"))

plt <- p1 | p2

.file <- fig.dir %&% "/Fig_Confounding_tSNE.pdf"
.gg.save(filename = .file, plot = plt, width=6, height=4)
print(plt)

PDF

Global statistics

.file <- fig.dir %&% "/Fig_histogram_pvalue.pdf"
pdf(.file, width = 4, height = 4)
hist(stat.avg.tab$pv.z, main="", xlab="gene-level p-value")
dev.off()

## quartz_off_screen 
##                 2

hist(stat.avg.tab$pv.z, main="", xlab="gene-level p-value")

PDF

Marginal statistics without confounding factor adjustment

gene.obs.assoc.dt <- 
    stat.tab[, .(mu=sum(tot)), by=.(gene, col)] %>%
    spread(key=col, value=mu, fill=0) %>% 
    (function(x) log(1 + x[, -"gene"])) %>%
    (function(x) scale(t(as.matrix(x)))) %>%
    (function(x) { x[is.na(x)] <- 0; x }) %>%
    test.gene.association %>%
    na.omit %>%
    rename(obs.beta = beta, obs.se = se, obs.pv = p.val) %>%
    as.data.table

gene.assoc.dt <- stat.avg.tab %>% na.omit %>%
    merge(gene.ontology$coding, by.x = "gene", by.y = "hgnc_symbol")

gene.assoc.dt <- gene.assoc.dt[order(gene.assoc.dt$pv.z),
                               head(.SD, 1),
                               by=.(gene, celltype)]

gene.assoc.dt <- gene.assoc.dt %>% 
    left_join(gene.obs.assoc.dt[pheno=="pathoAD"], by = c("gene", "celltype")) %>% 
    left_join(gene.cf.assoc.dt[pheno=="pathoAD"], by = c("gene", "celltype", "pheno"))

PV.CUTOFF <- 0.05/nrow(gene.assoc.dt)

plot.genome.view <- function(ct, .tot.dt, .cutoff){

    .chr.names <- c(as.character(1:22), "X", "Y", "MT")

    .dt <- .tot.dt[celltype == ct & chromosome_name %in% .chr.names] %>%
        mutate(chr = factor(chromosome_name, .chr.names)) %>% 
        mutate(z.trunc = pmin(pmax(z, -4), 4))

    .dt.show <- .dt[pv.z < .cutoff & abs(ln.mu) > .1]

    .dt.show.lab <- .dt.show[order(.dt$pv.z), head(.SD, 3), by=.(chr)] %>%
        filter(nchar(gene) > 0) %>% 
        na.omit

    x.lab.fn <- function(x) num.int(x/1e6)
    y.lab.fn <- function(x) num.sci(10^(-x))

    .aes <- aes(x=transcript_start,
                y=-log10(pmax(pv.z, 1e-30)),
                fill=z.trunc)

    ret <-
        .gg.plot(.dt, .aes) +
        ggtitle(ct) +
        theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1, size = 5)) +
        theme(title = element_text(size=8)) +
        theme(strip.text = element_text(size=6)) +
        facet_grid(. ~ chr, scales="free", space="free") +
        geom_hline(yintercept=-log10(.cutoff), size=.1, col="red") +
        scale_x_continuous(labels=x.lab.fn) +
        scale_y_continuous(labels=y.lab.fn) +
        ylab("p-value") + xlab("genomic locations (mb)") +
        scale_fill_gradient2(low="blue", high="red", guide=FALSE) +
        geom_point(pch=19, col="gray60", size=.5)

    if(nrow(.dt.show) > 0){
        ret <- ret + geom_point(data = .dt.show, pch=21)
    }

    if(nrow(.dt.show.lab) > 0){
        ret <- ret +
            ggrepel::geom_text_repel(data = .dt.show.lab, aes(label = gene), size = 1)
    }
    return(ret)
}

.ct.valid <- gene.assoc.dt[pv.z < PV.CUTOFF & abs(ln.mu) > .1] %>%
    na.omit %>%
    select(celltype) %>%
    unique %>%
    unlist

p.list <- 
    setdiff(.ct.valid, c("Per","Endo")) %>% 
    lapply(plot.genome.view, .tot.dt = gene.assoc.dt, .cutoff = PV.CUTOFF)

plt <- wrap_plots(p.list, ncol = 1)

.file <- fig.dir %&% "/Fig_Genome_View.pdf"
.gg.save(filename = .file, plot = plt, width=8, height=8)
print(plt)

PDF

Local statistics

Significant genes

.cell.count <- .fread("example/cfa/input.annot.gz") %>%
    (function(x) { colnames(x) <- c("TAG", "celltype"); x }) %>%
    (function(x) x[, .(Cell = .N), by=.(celltype)]) %>% 
    (function(x) x[, p.Cell := 100*Cell/sum(Cell)])

.gene.count <- 
    .gene.table[, .(Gene = .N), by = .(celltype)] %>% 
    (function(x) x[, p.Gene := 100*Gene/sum(Gene)])

.aes.cell <- aes(y=celltype, yend=celltype, x=0, xend=-Cell, colour = celltype, label=num.int(Cell) %&% " (" %&% num.round(p.Cell) %&% "%)")

.aes.gene <- aes(y=celltype, yend=celltype, x=0, xend=Gene, colour = celltype, label=num.int(Gene) %&% " (" %&% num.round(p.Gene) %&% "%)")

.plt <- function(.df, .aes) {
    .gg.plot(.df, .aes) +
    theme(axis.ticks.y = element_blank()) +
    theme(axis.title.y = element_blank()) +
    xlab("number of cells") +
    geom_segment(size=10) +
    ggrepel::geom_text_repel(size=2.5, colour = "gray20") +
    scale_x_continuous(labels=function(x) num.int(abs(x))) +
    scale_colour_brewer(palette = "Paired", guide = FALSE)
}

plt <- ((.plt(.cell.count, .aes.cell) + scale_y_discrete(position="right")) |
        .plt(.gene.count, .aes.gene))

.file <- fig.dir %&% "/Fig_Num_Discovery.pdf"
.gg.save(filename = .file, plot = plt, width=4, height=3)
print(plt)

PDF

celltype	gene	N	Chr	TSS	TES	FC	SD	Z	P
Astro	GFAP	1,609	17	44,903	44,917	1.38	0.03	12.86	7.7e-38
Astro	CHI3L1	687	1	203,178	203,187	1.50	0.06	10.85	2.0e-27
Astro	MT2A	2,899	16	56,608	56,610	1.19	0.02	9.27	1.9e-20
Astro	PRKG1	2,617	10	50,990	52,299	1.19	0.02	9.14	6.2e-20
Astro	HSP90AA1	1,342	14	102,080	102,140	1.27	0.03	8.92	4.6e-19
Astro	DPP10	3,236	2	114,442	115,846	1.16	0.02	8.64	5.8e-18
Astro	NEBL	3,438	10	20,779	21,294	1.15	0.02	8.44	3.2e-17
Astro	CPE	3,396	4	165,361	165,499	1.15	0.02	8.32	9.0e-17
Astro	AQP1	208	7	30,911	30,926	1.66	0.11	7.66	1.8e-14
Astro	ADCY2	2,802	5	7,396	7,831	1.15	0.02	7.58	3.5e-14
Astro	TRPM3	2,971	9	70,529	71,447	1.15	0.02	7.53	4.9e-14
Astro	CLU	4,400	8	27,596	27,615	1.11	0.02	7.10	1.2e-12
Astro	SNED1	303	2	240,998	241,096	1.48	0.08	7.01	2.5e-12
Astro	RGS6	286	14	71,932	72,567	1.48	0.08	6.79	1.1e-11
Astro	GJA1	2,458	6	121,435	121,450	1.14	0.02	6.71	1.9e-11
Astro	DST	3,121	6	56,457	56,955	1.13	0.02	6.68	2.4e-11
Astro	HSPB1	650	7	76,302	76,305	1.28	0.05	6.46	1.0e-10
Astro	RPS25	363	11	119,015	119,019	1.38	0.07	6.36	2.1e-10
Astro	ITGB4	499	17	75,721	75,758	1.32	0.06	6.33	2.5e-10
Astro	FYN	1,939	6	111,660	111,874	1.15	0.03	6.24	4.3e-10
Astro	RASGEF1B	1,827	4	81,426	82,045	1.15	0.03	6.09	1.1e-09
Astro	CD44	184	11	35,138	35,233	1.52	0.11	5.98	2.2e-09
Astro	LINGO1	1,706	15	77,613	77,821	1.15	0.03	5.81	6.2e-09
Astro	VCAN	242	5	83,471	83,583	1.43	0.09	5.74	9.4e-09
Astro	HSPA1A	559	6	31,815	31,818	1.26	0.05	5.61	2.0e-08
Astro	RFX4	1,641	12	106,583	106,763	1.15	0.03	5.60	2.2e-08
Astro	PTGDS	2,893	9	136,975	136,982	1.11	0.02	5.52	3.5e-08
Astro	CRYAB	500	11	111,908	111,924	1.27	0.06	5.49	4.0e-08
Astro	MAPK4	1,034	18	50,560	50,732	1.18	0.04	5.44	5.2e-08
Astro	PLEKHA5	562	12	19,129	19,377	1.25	0.05	5.43	5.5e-08
Astro	GPM6B	2,651	X	13,770	13,939	1.11	0.02	5.42	5.9e-08
Astro	FAM171A1	888	10	15,211	15,372	1.20	0.04	5.41	6.4e-08
Astro	PLEC	442	8	143,915	143,977	1.28	0.06	5.31	1.1e-07
Astro	CNTN1	1,818	12	40,692	41,073	1.13	0.03	5.30	1.2e-07
Astro	AEBP1	318	7	44,104	44,115	1.33	0.07	5.29	1.2e-07
Astro	TPT1	335	13	45,333	45,342	1.32	0.07	5.28	1.3e-07
Astro	MT1E	1,330	16	56,625	56,628	1.15	0.03	5.25	1.5e-07
Astro	ITPKB	1,209	1	226,631	226,740	1.16	0.03	5.22	1.8e-07
Astro	FLRT2	1,319	14	85,530	85,655	1.15	0.03	5.14	2.7e-07
Astro	TNC	137	9	115,019	115,119	1.51	0.12	5.14	2.8e-07
Astro	SYTL4	209	X	100,674	100,733	1.40	0.09	5.08	3.7e-07
Astro	RHOBTB3	1,335	5	95,713	95,825	1.15	0.03	5.06	4.2e-07
Astro	PARD3	2,625	10	34,109	34,816	1.10	0.02	5.03	4.9e-07
Astro	BAZ2B	1,427	2	159,318	159,617	1.14	0.03	5.01	5.5e-07
Astro	SON	1,390	21	33,543	33,578	1.14	0.03	4.97	6.5e-07
Astro	PDE4DIP	2,163	1	148,808	149,049	1.11	0.02	4.96	6.9e-07
Astro	MAML2	2,319	11	95,976	96,344	1.11	0.02	4.92	8.6e-07
Astro	LPIN1	863	2	11,677	11,828	1.18	0.04	4.90	9.5e-07
Astro	TENM4	396	11	78,652	79,442	1.27	0.06	4.89	1.0e-06
Endo	B2M	233	15	44,711	44,719	2.88	0.18	16.74	6.7e-63
Endo	MT2A	154	16	56,608	56,610	2.56	0.20	12.35	5.1e-35
Endo	TMSB10	156	2	84,905	84,907	2.46	0.19	11.86	1.9e-32
Endo	CLDN5	130	22	19,523	19,528	2.59	0.21	11.57	5.6e-31
Endo	IFITM3	107	11	319	328	2.39	0.21	9.76	1.6e-22
Endo	HLA-B	102	6	31,269	31,358	2.34	0.21	9.33	1.1e-20
Endo	BSG	87	19	571	584	2.36	0.23	8.80	1.4e-18
Endo	TMSB4X	129	X	12,975	12,978	2.06	0.17	8.79	1.5e-18
Endo	TSC22D1	83	13	44,432	44,578	2.34	0.23	8.52	1.5e-17
Endo	FLT1	70	13	28,300	28,496	2.40	0.26	8.22	2.0e-16
Endo	APOD	93	3	195,568	195,585	2.14	0.20	7.99	1.3e-15
Endo	RPL13	88	16	89,560	89,565	2.05	0.20	7.37	1.8e-13
Endo	RASGEF1B	83	4	81,426	82,045	2.05	0.20	7.23	4.8e-13
Endo	RPS19	84	19	41,860	41,873	2.05	0.20	7.23	4.9e-13
Endo	LINGO1	77	15	77,613	77,821	2.05	0.21	7.02	2.2e-12
Endo	IGFBP7	72	4	57,030	57,111	2.09	0.22	6.99	2.8e-12
Endo	RBMS3	76	3	28,574	30,011	2.05	0.21	6.98	2.9e-12
Endo	HLA-C	67	6	31,268	31,273	2.12	0.23	6.95	3.7e-12
Endo	DLC1	62	8	13,083	13,605	2.13	0.24	6.75	1.5e-11
Endo	ACTB	86	7	5,526	5,564	1.91	0.19	6.62	3.6e-11
Endo	PTMA	74	2	231,706	231,714	1.98	0.21	6.56	5.5e-11
Endo	DUSP1	53	5	172,768	172,772	2.17	0.26	6.51	7.5e-11
Endo	HLA-A	65	6	29,941	29,946	2.03	0.22	6.46	1.0e-10
Endo	EEF1A1	86	6	73,515	73,524	1.88	0.18	6.45	1.1e-10
Endo	RPS15	72	19	1,438	1,441	1.95	0.21	6.31	2.7e-10
Endo	HSP90AA1	82	14	102,080	102,140	1.87	0.19	6.26	3.9e-10
Endo	EIF1	65	17	41,688	41,693	1.96	0.22	6.14	8.3e-10
Endo	RPL13A	69	19	49,487	49,493	1.92	0.21	6.08	1.2e-09
Endo	EPAS1	54	2	46,293	46,387	2.04	0.24	6.06	1.4e-09
Endo	RPL10	70	X	154,389	154,410	1.91	0.20	6.05	1.4e-09
Endo	ST6GALNAC3	55	1	76,074	76,635	2.03	0.24	6.03	1.6e-09
Endo	IFI27	61	14	94,104	94,117	1.96	0.22	5.96	2.4e-09
Endo	ABCB1	59	7	87,503	87,714	1.97	0.22	5.95	2.7e-09
Endo	PTPRG	66	3	61,561	62,298	1.89	0.21	5.81	6.1e-09
Endo	SPARCL1	68	4	87,473	87,532	1.80	0.19	5.44	5.2e-08
Endo	RPL41	53	12	56,116	56,118	1.89	0.22	5.38	7.5e-08
Endo	SYNE1	50	6	152,121	152,638	1.91	0.23	5.35	9.0e-08
Endo	HLA-E	53	6	30,489	30,495	1.87	0.22	5.25	1.5e-07
Endo	MBNL1	53	3	152,243	152,466	1.80	0.21	4.93	8.1e-07
Ex	MTRNR2L1	3,241	17	22,523	22,525	1.93	0.03	37.50	7.7e-308
Ex	DHFR	17,260	5	80,626	80,655	1.18	0.01	21.58	2.7e-103
Ex	MTRNR2L8	2,829	11	10,507	10,510	1.41	0.03	18.51	1.6e-76
Ex	ARL17B	4,977	17	46,274	46,362	0.79	0.01	-16.56	1.4e-61
Ex	ELK1	5,885	X	47,635	47,651	1.21	0.02	14.63	1.7e-48
Ex	EIF5A	2,775	17	7,306	7,313	0.77	0.01	-14.09	4.5e-45
Ex	NIPAL2	8,842	8	98,189	98,295	0.86	0.01	-14.02	1.2e-44
Ex	MTRNR2L12	4,632	3	96,617	96,619	1.21	0.02	12.92	3.4e-38
Ex	SPDYE2	3,447	7	102,551	102,563	0.80	0.01	-12.91	4.2e-38
Ex	SNCG	4,460	10	86,958	86,964	0.83	0.01	-12.22	2.3e-34
Ex	HSPA8	9,252	11	123,057	123,064	1.12	0.01	11.31	1.1e-29
Ex	NNAT	4,573	20	37,521	37,524	1.17	0.02	10.83	2.5e-27
Ex	P2RY14	5,427	3	151,212	151,279	1.16	0.02	10.73	7.7e-27
Ex	LRPAP1	5,929	4	3,503	3,533	1.14	0.01	10.16	2.9e-24
Ex	SSTR2	4,969	17	73,165	73,177	1.15	0.02	9.88	4.9e-23
Ex	STAT4	5,406	2	191,029	191,152	0.88	0.01	-9.83	8.6e-23
Ex	CNTN6	2,329	3	1,092	1,405	0.82	0.02	-9.81	1.1e-22
Ex	RPH3A	7,851	12	112,570	112,899	0.90	0.01	-9.61	7.1e-22
Ex	ITPR2	6,605	12	26,335	26,834	0.89	0.01	-9.59	8.4e-22
Ex	PDE5A	7,744	4	119,494	119,629	0.90	0.01	-9.55	1.3e-21
Ex	SEMA3A	6,486	7	83,955	84,493	0.89	0.01	-9.19	3.8e-20
Ex	OR2L13	5,594	1	248,095	248,102	1.13	0.02	9.15	5.5e-20
Ex	PTPN14	2,555	1	214,348	214,553	0.84	0.02	-8.96	3.2e-19
Ex	EEF1A2	3,296	20	63,488	63,500	1.16	0.02	8.78	1.6e-18
Ex	LRRC37A3	6,259	17	64,854	64,920	0.90	0.01	-8.36	6.1e-17
Ex	PTCHD4	2,993	6	47,878	48,069	0.86	0.02	-8.24	1.7e-16
Ex	PCED1B	1,681	12	47,079	47,237	0.83	0.02	-7.83	5.1e-15
Ex	EGR1	1,331	5	138,465	138,470	0.81	0.02	-7.73	1.1e-14
Ex	VGF	2,487	7	101,162	101,166	0.86	0.02	-7.71	1.3e-14
Ex	VSTM2L	4,594	20	37,903	37,946	0.89	0.01	-7.61	2.7e-14
Ex	CD81	4,281	11	2,376	2,398	1.12	0.02	7.47	8.0e-14
Ex	NBPF11	1,872	1	148,102	148,153	0.84	0.02	-7.36	1.9e-13
Ex	TEF	4,331	22	41,367	41,400	1.12	0.02	7.28	3.3e-13
Ex	LRRC37A2	1,636	17	46,511	46,556	0.84	0.02	-7.08	1.4e-12
Ex	TSC22D3	3,996	X	107,713	107,778	1.12	0.02	7.08	1.4e-12
Ex	ADAMTS2	1,893	5	179,110	179,346	1.18	0.03	7.05	1.8e-12
Ex	CNGB1	3,271	16	57,882	57,972	1.13	0.02	6.98	3.0e-12
Ex	MSH4	1,641	1	75,796	75,914	1.19	0.03	6.95	3.6e-12
Ex	PLA2G4A	732	1	186,828	186,989	0.78	0.03	-6.94	3.8e-12
Ex	SLC38A2	4,492	12	46,358	46,373	1.11	0.02	6.89	5.4e-12
Ex	FOS	89	14	75,278	75,283	0.51	0.05	-6.89	5.6e-12
Ex	ST18	368	8	52,110	52,461	0.70	0.04	-6.88	6.2e-12
Ex	OTOGL	1,513	12	80,099	80,381	0.84	0.02	-6.83	8.8e-12
Ex	RASD2	1,808	22	35,540	35,554	1.17	0.03	6.55	5.7e-11
Ex	DNAJB1	3,696	19	14,514	14,530	1.11	0.02	6.46	1.1e-10
Ex	HRH3	1,074	20	62,214	62,221	1.21	0.04	6.36	2.0e-10
Ex	GPS2	1,668	17	7,311	7,316	1.17	0.03	6.30	2.9e-10
Ex	DUSP26	3,574	8	33,591	33,601	1.11	0.02	6.27	3.7e-10
Ex	DUSP6	696	12	89,347	89,353	0.79	0.03	-6.17	6.7e-10
Ex	PMPCB	3,522	7	103,297	103,330	1.11	0.02	6.12	9.3e-10
Ex	NPTX2	1,317	7	98,617	98,630	0.85	0.02	-6.10	1.1e-09
Ex	KCNV2	875	9	2,717	2,731	1.23	0.04	6.07	1.3e-09
Ex	HLA-DRB1	184	6	32,578	32,590	0.66	0.05	-5.96	2.5e-09
Ex	MSMO1	3,701	4	165,327	165,344	1.10	0.02	5.90	3.6e-09
Ex	PHLDB2	466	3	111,732	111,977	0.77	0.03	-5.82	5.7e-09
Ex	PCSK6	1,238	15	101,297	101,526	0.85	0.02	-5.81	6.1e-09
Ex	PILRB	2,763	7	100,352	100,368	0.90	0.02	-5.79	7.2e-09
Ex	RPRM	1,069	2	153,477	153,479	1.19	0.04	5.75	9.1e-09
Ex	MYOM2	2,043	8	2,045	2,166	0.88	0.02	-5.73	1.0e-08
Ex	YIPF6	2,913	X	68,498	68,538	1.11	0.02	5.71	1.1e-08
Ex	LAMP5	3,370	20	9,514	9,531	1.10	0.02	5.71	1.1e-08
Ex	TSPAN17	1,337	5	176,647	176,660	1.17	0.03	5.70	1.2e-08
Ex	IL34	1,963	16	70,579	70,661	1.14	0.03	5.68	1.4e-08
Ex	SYT6	340	1	114,089	114,154	1.35	0.07	5.67	1.4e-08
Ex	TMEM233	1,512	12	119,593	119,644	1.16	0.03	5.64	1.7e-08
Ex	IL1RAP	1,726	3	190,514	190,660	0.87	0.02	-5.59	2.2e-08
Ex	CTXN3	895	5	127,649	127,659	0.83	0.03	-5.56	2.6e-08
Ex	TIMP2	3,291	17	78,852	78,926	1.10	0.02	5.54	3.0e-08
Ex	ZDHHC2	2,055	8	17,156	17,225	0.89	0.02	-5.52	3.4e-08
Ex	CD24	583	6	106,969	106,976	1.25	0.05	5.52	3.4e-08
Ex	DUSP8	2,317	11	1,554	1,573	1.12	0.02	5.50	3.8e-08
Ex	IGFBP5	3,198	2	216,672	216,696	1.10	0.02	5.47	4.6e-08
Ex	SMAD4	2,183	18	51,028	51,086	1.12	0.02	5.43	5.5e-08
Ex	NR4A1	788	12	52,022	52,060	0.83	0.03	-5.39	7.2e-08
Ex	CD53	882	1	110,871	110,900	0.84	0.03	-5.37	7.8e-08
Ex	TMED2	2,832	12	123,584	123,599	1.11	0.02	5.36	8.5e-08
Ex	TCF7L2	2,481	10	112,950	113,168	0.90	0.02	-5.34	9.2e-08
Ex	HUNK	1,723	21	31,873	32,045	0.88	0.02	-5.29	1.2e-07
Ex	TMEM144	785	4	158,201	158,256	0.83	0.03	-5.28	1.3e-07
Ex	PPAT	1,939	4	56,393	56,436	1.13	0.03	5.26	1.5e-07
Ex	CFAP61	1,540	20	20,052	20,361	0.88	0.02	-5.24	1.6e-07
Ex	GRIK1	1,903	21	29,536	29,941	0.89	0.02	-5.22	1.8e-07
Ex	TTF2	1,118	1	117,060	117,108	0.86	0.03	-5.21	1.9e-07
Ex	GRIK3	2,112	1	36,795	37,035	1.12	0.02	5.19	2.1e-07
Ex	SPTSSB	1,285	3	161,344	161,373	0.87	0.02	-5.14	2.8e-07
Ex	MTRNR2L6	198	7	142,666	142,668	1.42	0.10	5.12	3.0e-07
Ex	COBLL1	1,434	2	164,653	164,844	0.88	0.02	-5.06	4.2e-07
Ex	PLD1	931	3	171,600	171,811	0.85	0.03	-5.06	4.2e-07
Ex	HTR4	2,094	5	148,451	148,678	0.90	0.02	-5.06	4.3e-07
Ex	MKNK2	1,395	19	2,037	2,052	1.14	0.03	5.04	4.7e-07
Ex	ZDHHC22	2,654	14	77,131	77,143	1.10	0.02	4.98	6.4e-07
Ex	MYO1E	773	15	59,132	59,373	0.84	0.03	-4.96	6.9e-07
Ex	CBFB	1,309	16	67,028	67,102	1.15	0.03	4.95	7.3e-07
Ex	COMMD2	2,506	3	149,738	149,753	1.10	0.02	4.94	7.9e-07
Ex	NEDD9	923	6	11,183	11,383	0.85	0.03	-4.92	8.9e-07
Ex	CREG1	2,367	1	167,529	167,554	1.11	0.02	4.90	9.7e-07
In	MTRNR2L1	732	17	22,523	22,525	1.83	0.07	16.59	7.7e-62
In	DHFR	4,437	5	80,626	80,655	1.20	0.02	12.16	4.9e-34
In	RASGEF1B	6,371	4	81,426	82,045	1.13	0.01	9.71	2.8e-22
In	MTRNR2L8	761	11	10,507	10,510	1.40	0.05	9.36	8.1e-21
In	SLC26A3	3,477	7	107,765	107,804	1.17	0.02	9.32	1.1e-20
In	IQCJ-SCHIP1	4,990	3	158,962	159,898	1.12	0.02	8.21	2.1e-16
In	PCSK1N	5,724	X	48,831	48,836	1.11	0.01	7.89	3.1e-15
In	LINGO1	5,398	15	77,613	77,821	1.10	0.02	7.33	2.3e-13
In	ARL17B	805	17	46,274	46,362	0.78	0.03	-7.25	4.0e-13
In	CLU	2,787	8	27,596	27,615	1.15	0.02	7.17	7.3e-13
In	MTRNR2L12	1,121	3	96,617	96,619	1.22	0.04	6.62	3.7e-11
In	EIF5A	369	17	7,306	7,313	0.72	0.04	-6.54	6.2e-11
In	FTL	4,234	19	48,965	48,967	1.10	0.02	6.22	5.0e-10
In	CENPP	626	9	92,325	92,621	0.80	0.03	-5.75	8.8e-09
In	SNCG	1,700	10	86,958	86,964	0.88	0.02	-5.23	1.7e-07
Microglia	SPP1	1,391	4	87,975	87,984	1.58	0.04	17.12	1.0e-65
Microglia	PLXDC2	2,931	10	19,816	20,290	1.28	0.02	13.26	3.8e-40
Microglia	SLC1A3	1,513	5	36,606	36,689	1.37	0.04	12.38	3.4e-35
Microglia	RASGEF1B	439	4	81,426	82,045	1.59	0.07	9.92	3.5e-23
Microglia	APOE	842	19	44,905	44,910	1.39	0.05	9.67	4.0e-22
Microglia	MBNL1	737	3	152,243	152,466	1.39	0.05	9.11	8.0e-20
Microglia	DOCK8	1,064	9	214	466	1.31	0.04	8.80	1.4e-18
Microglia	HS3ST4	1,026	16	25,691	26,138	1.31	0.04	8.65	5.1e-18
Microglia	FKBP5	701	6	35,573	35,729	1.38	0.05	8.64	5.5e-18
Microglia	FRMD4A	2,031	10	13,643	14,463	1.21	0.03	8.63	5.9e-18
Microglia	RPS19	519	19	41,860	41,873	1.44	0.06	8.51	1.7e-17
Microglia	CACNA1A	302	19	13,206	13,634	1.60	0.09	8.45	3.0e-17
Microglia	ARHGAP24	1,230	4	85,475	86,003	1.26	0.04	8.30	1.1e-16
Microglia	FTL	986	19	48,965	48,967	1.29	0.04	7.97	1.5e-15
Microglia	DOCK10	545	2	224,765	225,043	1.39	0.06	7.89	3.0e-15
Microglia	MEF2C	1,291	5	88,717	88,905	1.24	0.03	7.74	9.9e-15
Microglia	HDAC9	812	7	18,086	19,003	1.31	0.05	7.73	1.0e-14
Microglia	UBE2E2	660	3	23,203	23,592	1.34	0.05	7.55	4.5e-14
Microglia	FOXN3	1,091	14	89,124	89,620	1.25	0.04	7.52	5.5e-14
Microglia	FMN1	439	15	32,765	33,195	1.42	0.07	7.51	5.8e-14
Microglia	ELMO1	974	7	36,854	37,450	1.27	0.04	7.49	7.0e-14
Microglia	LHFPL2	507	5	78,485	78,771	1.38	0.06	7.31	2.7e-13
Microglia	CSGALNACT1	558	8	19,404	19,759	1.35	0.06	7.24	4.3e-13
Microglia	DOCK4	1,808	7	111,726	112,207	1.18	0.03	7.24	4.4e-13
Microglia	GRID2	588	4	92,303	93,811	1.34	0.05	7.22	5.1e-13
Microglia	RPL13	391	16	89,560	89,565	1.42	0.07	7.16	8.3e-13
Microglia	LINGO1	383	15	77,613	77,821	1.43	0.07	7.15	8.8e-13
Microglia	MAML2	977	11	95,976	96,344	1.25	0.04	7.02	2.2e-12
Microglia	PICALM	635	11	85,957	86,070	1.31	0.05	6.96	3.4e-12
Microglia	BHLHE41	310	12	26,120	26,126	1.47	0.08	6.96	3.5e-12
Microglia	VSIG4	303	X	66,021	66,041	1.47	0.08	6.92	4.4e-12
Microglia	FGD4	512	12	32,399	32,647	1.35	0.06	6.91	4.9e-12
Microglia	TBC1D5	504	3	17,157	18,445	1.35	0.06	6.90	5.3e-12
Microglia	CSF1R	423	5	150,053	150,114	1.38	0.07	6.82	8.9e-12
Microglia	MEF2A	1,099	15	99,565	99,717	1.22	0.04	6.63	3.3e-11
Microglia	LAPTM5	293	1	30,732	30,758	1.46	0.08	6.62	3.6e-11
Microglia	CELF2	887	10	10,798	11,337	1.25	0.04	6.62	3.7e-11
Microglia	SNCA	216	4	89,700	89,839	1.54	0.10	6.60	4.0e-11
Microglia	TMSB4X	855	X	12,975	12,978	1.25	0.04	6.55	5.8e-11
Microglia	SLC9A9	763	3	143,265	143,849	1.26	0.05	6.49	8.4e-11
Microglia	GLDN	306	15	51,341	51,409	1.43	0.08	6.43	1.2e-10
Microglia	SLC26A3	199	7	107,765	107,804	1.55	0.11	6.43	1.3e-10
Microglia	RUNX1	635	21	34,787	36,005	1.28	0.05	6.40	1.5e-10
Microglia	GNAQ	670	9	77,716	78,032	1.28	0.05	6.38	1.7e-10
Microglia	OXR1	696	8	106,359	106,753	1.27	0.05	6.36	2.0e-10
Microglia	CYFIP1	372	15	22,867	22,982	1.38	0.07	6.30	3.0e-10
Microglia	ANKRD44	748	2	196,967	197,312	1.25	0.05	6.21	5.4e-10
Microglia	PSAP	390	10	71,816	71,852	1.36	0.07	6.17	7.0e-10
Microglia	SSH2	883	17	29,625	29,931	1.23	0.04	6.12	9.3e-10
Microglia	NHSL1	204	6	138,422	138,693	1.51	0.10	6.11	9.7e-10
Microglia	RPLP1	286	15	69,452	69,457	1.42	0.08	6.09	1.1e-09
Microglia	SH3RF3	515	2	109,129	109,505	1.30	0.06	6.08	1.2e-09
Microglia	APOC1	125	19	44,914	44,920	1.66	0.14	6.08	1.2e-09
Microglia	TBXAS1	664	7	139,777	140,021	1.26	0.05	6.07	1.3e-09
Microglia	LDLRAD4	867	18	13,217	13,653	1.23	0.04	6.07	1.3e-09
Microglia	APBB1IP	997	10	26,438	26,568	1.21	0.04	5.96	2.5e-09
Microglia	ZFHX3	511	16	72,782	73,892	1.29	0.06	5.94	2.8e-09
Microglia	ADGRG1	235	16	57,610	57,666	1.45	0.09	5.94	2.8e-09
Microglia	SKAP2	363	7	26,667	26,996	1.35	0.07	5.92	3.2e-09
Microglia	DPYD	322	1	97,077	97,995	1.38	0.07	5.92	3.2e-09
Microglia	SLTM	163	15	58,879	58,934	1.55	0.12	5.90	3.6e-09
Microglia	MAN2A1	151	5	109,689	109,870	1.56	0.12	5.81	6.4e-09
Microglia	JAZF1	411	7	27,830	28,181	1.32	0.06	5.79	7.0e-09
Microglia	ALCAM	166	3	105,366	105,577	1.53	0.11	5.79	7.2e-09
Microglia	CD81	192	11	2,376	2,398	1.48	0.10	5.71	1.1e-08
Microglia	RPL28	222	19	55,385	55,404	1.44	0.09	5.69	1.2e-08
Microglia	ST6GAL1	1,078	3	186,930	187,079	1.19	0.04	5.64	1.7e-08
Microglia	EPB41L2	797	6	130,839	131,064	1.22	0.04	5.62	2.0e-08
Microglia	STARD13	157	13	33,103	33,351	1.53	0.12	5.60	2.1e-08
Microglia	ITPR2	1,031	12	26,335	26,834	1.19	0.04	5.57	2.6e-08
Microglia	ZFP36L1	406	14	68,787	68,797	1.31	0.06	5.56	2.6e-08
Microglia	RPL13A	255	19	49,487	49,493	1.40	0.08	5.54	3.0e-08
Microglia	TANC2	247	17	63,009	63,428	1.40	0.09	5.51	3.5e-08
Microglia	C1QC	320	1	22,643	22,649	1.35	0.07	5.51	3.6e-08
Microglia	ZEB2	620	2	144,364	144,522	1.24	0.05	5.51	3.7e-08
Microglia	TAB2	378	6	149,218	149,412	1.32	0.07	5.50	3.7e-08
Microglia	ACTB	334	7	5,526	5,564	1.34	0.07	5.47	4.4e-08
Microglia	HSP90AA1	420	14	102,080	102,140	1.30	0.06	5.46	4.9e-08
Microglia	PNISR	346	6	99,398	99,426	1.33	0.07	5.43	5.7e-08
Microglia	AUTS2	319	7	69,598	70,794	1.34	0.07	5.41	6.5e-08
Microglia	EIF4G3	414	1	20,806	21,177	1.30	0.06	5.39	7.2e-08
Microglia	SRGAP2	956	1	206,203	206,465	1.19	0.04	5.38	7.4e-08
Microglia	HLA-B	231	6	31,269	31,358	1.40	0.09	5.34	9.1e-08
Microglia	IFNGR1	254	6	137,197	137,220	1.37	0.08	5.24	1.6e-07
Microglia	LPCAT2	389	16	55,509	55,587	1.30	0.06	5.23	1.7e-07
Microglia	P2RY12	357	3	151,336	151,385	1.31	0.07	5.21	1.9e-07
Microglia	RPS27A	203	2	55,231	55,236	1.42	0.10	5.19	2.1e-07
Microglia	N4BP2L2	382	13	32,432	32,539	1.29	0.06	5.16	2.4e-07
Microglia	FOXP1	349	3	70,952	71,584	1.31	0.07	5.16	2.4e-07
Microglia	SPIDR	287	8	47,260	47,737	1.34	0.08	5.16	2.4e-07
Microglia	SHTN1	290	10	116,881	117,127	1.34	0.08	5.13	2.9e-07
Microglia	CD74	709	5	150,400	150,413	1.21	0.04	5.09	3.5e-07
Microglia	BNC2	311	9	16,409	16,871	1.32	0.07	5.09	3.5e-07
Microglia	TCF12	579	15	56,918	57,300	1.23	0.05	5.09	3.6e-07
Microglia	RAP1A	295	1	111,542	111,717	1.33	0.08	5.07	3.9e-07
Microglia	DTNA	193	18	34,493	34,892	1.42	0.10	5.07	3.9e-07
Microglia	MED13L	553	12	115,957	116,278	1.24	0.05	5.07	4.0e-07
Microglia	SLC8A1	799	2	40,097	40,612	1.19	0.04	5.06	4.2e-07
Microglia	MYO1E	155	15	59,132	59,373	1.47	0.11	5.05	4.4e-07
Microglia	SORL1	688	11	121,452	121,634	1.21	0.05	5.03	4.8e-07
Microglia	ARHGAP15	658	2	143,091	143,769	1.21	0.05	5.03	5.0e-07
Microglia	ANKRD12	208	18	9,136	9,286	1.40	0.09	5.01	5.4e-07
Microglia	TMSB10	236	2	84,905	84,907	1.37	0.09	5.01	5.5e-07
Microglia	MAML3	639	4	139,716	140,155	1.22	0.05	5.01	5.6e-07
Microglia	ATM	333	11	108,222	108,370	1.31	0.07	5.00	5.7e-07
Microglia	SAMD12	137	8	118,189	118,623	1.49	0.12	4.97	6.8e-07
Microglia	SIK3	246	11	116,843	117,099	1.36	0.08	4.96	7.0e-07
Microglia	ZFP36L2	307	2	43,222	43,227	1.32	0.07	4.96	7.2e-07
Microglia	WWOX	404	16	78,099	79,213	1.27	0.06	4.94	7.9e-07
Microglia	TNRC6B	343	22	40,044	40,336	1.30	0.07	4.94	7.9e-07
Microglia	DIAPH2	595	X	96,684	97,605	1.22	0.05	4.93	8.0e-07
Microglia	FHIT	567	3	59,747	61,252	1.23	0.05	4.93	8.2e-07
Microglia	C3	500	19	6,677	6,731	1.24	0.05	4.92	8.7e-07
Microglia	ARMC9	100	2	231,198	231,377	1.57	0.14	4.89	1.0e-06
Microglia	GPATCH2L	150	14	76,151	76,255	1.46	0.11	4.89	1.0e-06
Oligo	NTM	7,395	11	131,370	132,337	1.15	0.01	11.70	1.2e-31
Oligo	ARL17B	382	17	46,274	46,362	0.60	0.03	-10.22	1.6e-24
Oligo	HSPA1B	461	6	31,827	31,831	0.63	0.03	-10.06	8.2e-24
Oligo	TLE4	3,345	9	79,571	79,727	1.14	0.02	7.82	5.5e-15
Oligo	PRKX	573	X	3,604	3,714	1.36	0.06	7.39	1.5e-13
Oligo	CHRM5	625	15	33,968	34,068	0.75	0.03	-7.16	8.1e-13
Oligo	ABCA6	651	17	69,078	69,142	1.30	0.05	6.76	1.4e-11
Oligo	PTPRM	831	18	7,566	8,407	1.26	0.04	6.66	2.7e-11
Oligo	ADGRL3	2,857	4	61,201	62,079	0.88	0.02	-6.59	4.4e-11
Oligo	ZNF365	2,198	10	62,374	62,481	1.15	0.02	6.50	8.1e-11
Oligo	KCNIP4	1,472	4	20,728	21,949	1.18	0.03	6.38	1.8e-10
Oligo	ADAMTS18	618	16	77,247	77,436	1.27	0.05	6.11	9.9e-10
Oligo	HSD11B1	375	1	209,686	209,735	0.74	0.04	-5.93	3.0e-09
Oligo	LURAP1L	457	9	12,775	12,824	1.31	0.06	5.89	3.9e-09
Oligo	FKBP5	1,885	6	35,573	35,729	1.14	0.03	5.88	4.0e-09
Oligo	RHOBTB1	1,289	10	60,869	61,002	1.17	0.03	5.65	1.6e-08
Oligo	CPQ	2,658	8	96,645	97,150	1.11	0.02	5.61	2.0e-08
Oligo	LAMA2	2,688	6	128,883	129,517	1.11	0.02	5.52	3.4e-08
Oligo	PDK4	1,037	7	95,583	95,597	1.18	0.04	5.46	4.9e-08
Oligo	CARNS1	1,529	11	67,414	67,426	0.87	0.02	-5.30	1.2e-07
Oligo	LIFR	1,840	5	38,474	38,609	1.13	0.03	5.26	1.4e-07
Oligo	HSPA1A	2,228	6	31,815	31,818	0.90	0.02	-5.23	1.7e-07
Oligo	MDGA2	1,052	14	46,839	47,676	1.17	0.04	5.07	4.0e-07
Oligo	MT3	1,287	16	56,589	56,592	1.15	0.03	5.04	4.6e-07
Oligo	NKX6-2	1,677	10	132,783	132,787	1.13	0.03	5.04	4.6e-07
Oligo	PTPN13	1,448	4	86,594	86,816	1.14	0.03	5.01	5.3e-07
Oligo	DCC	855	18	52,340	53,536	1.18	0.04	4.99	6.1e-07
Oligo	SORCS2	851	4	7,192	7,743	1.18	0.04	4.99	6.1e-07
Oligo	TTC3	1,825	21	37,073	37,204	1.12	0.03	4.90	9.7e-07
OPC	DSCAM	8,292	21	40,010	40,848	1.12	0.01	10.30	6.8e-25
OPC	LRRC4C	7,942	11	40,114	41,460	1.10	0.01	8.86	8.2e-19
OPC	LRP1B	7,092	2	140,231	142,132	1.10	0.01	8.17	3.0e-16
OPC	NOVA1	3,192	14	26,443	26,599	1.15	0.02	7.95	1.9e-15
OPC	NLGN1	5,979	3	173,396	174,287	1.11	0.01	7.75	9.1e-15
OPC	PTPRZ1	4,661	7	121,873	122,063	1.12	0.02	7.70	1.3e-14
OPC	LINGO1	1,267	15	77,613	77,821	1.22	0.03	7.08	1.4e-12
OPC	VCAN	3,832	5	83,471	83,583	1.12	0.02	7.06	1.6e-12
OPC	NTM	5,256	11	131,370	132,337	1.10	0.02	6.97	3.1e-12
OPC	NRG3	2,130	10	81,875	82,988	1.16	0.02	6.80	1.1e-11
OPC	RASGEF1B	1,293	4	81,426	82,045	1.20	0.03	6.74	1.6e-11
OPC	MAML2	2,153	11	95,976	96,344	1.15	0.02	6.41	1.5e-10
OPC	FGF12	2,740	3	192,139	192,768	1.13	0.02	6.31	2.7e-10
OPC	OLIG1	1,459	21	33,070	33,073	1.18	0.03	6.23	4.6e-10
OPC	KAZN	2,738	1	13,892	15,119	1.12	0.02	5.89	3.8e-09
OPC	SLC26A3	550	7	107,765	107,804	1.26	0.05	5.54	3.1e-08
OPC	ATRNL1	2,230	10	115,093	115,949	1.12	0.02	5.32	1.0e-07
OPC	FHIT	2,169	3	59,747	61,252	1.12	0.02	5.23	1.7e-07
OPC	PPP2R2B	1,949	5	146,581	147,085	1.12	0.03	5.10	3.4e-07
OPC	NCAM1	2,059	11	112,961	113,279	1.12	0.02	5.01	5.4e-07
OPC	ZBTB20	2,699	3	114,314	115,148	1.10	0.02	5.01	5.5e-07
Per	HSP90AA1	164	14	102,080	102,140	4.66	0.35	20.75	1.1e-95

Counterfactual confounding factor adjustment (COCOA) in single-cell RNA-seq matrix

Yongjin Park

2020-11-03 19:12:42

0. Annotate cell types for all the cells

1. Extract cells with underlying phenotypes are annotated

2. Create another file set for COCOA analysis

3. Run the counterfactual adjustment model

First, show confounders and genes associated with the confounding effects

Global statistics

Marginal statistics without confounding factor adjustment

Local statistics

Significant genes